You are here

Beaver Creek Experimental Watershed: Research on fuel treatment effects

January, 2006

The Southwest Watershed Science Team is engaged in research on the Beaver Creek Experimental Watershed (BCEW) outside of Flagstaff, Arizona. There is over 20 years of hydrologic, climatic, vegetation, fuels, soils, and wildlife data from the 1950s through the 1980s which provide background for the study (see the project Beaver Creek Experimental Watershed: Research on forest management, streamflow, and forage production for details on historical research at BCEW). 

In the mid-2000's researchers reinstated research at BCEW to collect data on climate, stream flow, vegetation, forest floor, and soil conditions. The main goal of ongoing research at BCEW is to provide land managers with information about the ecological effects of fuel treatments in the ponderosa pine forests and pinyon-juniper woodlands at a watershed scale.


Study Site

Vegetation types found within the Beaver Creek Experimental Watershed.
Vegetation types found within the Beaver Creek Experimental Watershed.
The BCEW encompasses 111,375 hectares on the Coconino National Forest. The BCEW is upstream from the junction of Beaver Creek and the Verde River, and its part of the Salt-Verde River Basins.

The Beaver Creek watershed lies along the Mogollon Rim within the largest continuous stands of ponderosa pine in the United States--a belt of trees extending for nearly 321 km across Arizona. In ascending order of elevation, the three vegetation types found on the Beaver Creek watershed are semi-desert shrubs (<1,500 m elevation), pinyon-juniper (1,500-1,800 m), and ponderosa pine (>2,000 m).

Additional information on the BCEW is available online through the Beaver Creek Environmental Atlas maintained by Northern Arizona University.

Experimental Design

In 2006, scientists with RMRS and Northern Arizona University re-instated research in five watershed subunits at BCEW (watersheds 9, 11, 12, 13, and 14). The new research is focused on fuel treatment effects, specifically the impacts of burning and/or thinning on stream flow and water quality, vegetation, forest floor, and soil conditions.

Fuel Treatment Effects

The Coconino National Forest is working with the scientists to conduct thinning and burning treatments in five watershed subunits (Table 1). Vegetation and forest floor will be measured before and two years after treatment. Stream flow measurements will be compared between treated and untreated subunits. Continuously collected climate data will also inform experimental results.

This study will provide novel information on fuels treatment effects at a watershed scale. Such information is critical to inform fuels treatment in Arizona and the Southwest that combine thinning and burning to reduce wildfire hazard.

Table 1. Characteristics of watershed subunits on the Beaver Creek Experimental Watershed and description of experimental fuel treatments.
Watershed subunit Area (hectare) Fuel treatment Num. permanent plots Original measurement dates
9 454 Thin 22% of basal area, burn 33% 186 1965, 1975
11 76 Burn 100% 189 1965, 1975, 1993
12 184 Burn 100% 186 1962, 1975, 1992
13 368 Control 186 1965, 1975
14 546 Thin 55% of basal area, burn 100% 194 1965, 1975, 1994

Stream Flow and Water Quality

Streamflow measurements are based on stage height and flume ratings from supercritical, trapezoidal flumes in each watershed. Stream stage heights are converted to flow volumes using hydraulic rating formulas.

Samples for water quality analysis (suspended sediment and anions/cations) are collected at gauging flumes in each watershed during monsoon storm events and winter snowpack runoff.  Samples will be flow or time composited from ISCO discrete samplers depending on the flow characteristics for the period.


Ponderosa pine stand on the Beaver Creek Experimental Watershed.
Ponderosa pine stand on the Beaver Creek Experimental Watershed.
Overstory composition, understory, and forest floor will be determined on existing permanent plots. The methods described by Weatherspoon and McIver (2000) for the Fire and Fire Surrogate Study will be used as much as possible. The following methods used in the previous surveys (1960s, 1975, and mid-1990s) will also be repeated for consistency (Barger and Ffolliott 1969, Clary 1964):

  • Overstory measurements: point sampling using a basal area factor 25 prism.

  • Understory species: double sampling procedure in 0.88 m2 plots.

  • Understory biomass: clipped plots every 10 sampling points to help convert field measures of green weight to oven dry weight.

  • Cover of bare soil, woody debris, and rock: 10-m long line transects in each sample plot.

Forest Floor and Soils

The forest floor and soil conditions will be measured using the method of Ffolliott and others (1968), which is similar to the Fire and Fire Surrogate method (Weatherspoon and McIver 2000):

  • Weight of forest floor layers: oven-dry weights from 929 cm2 samples taken in one quadrant randomly selected from each set of 10 plots.

  • Litter depth: depth measurements (to the nearest 2.5 mm) at the four sides of the 929 cm2 litter samples.

  • Nutrient content of forest floor: three samples adjacent to each litter sample to determine the carbon, nitrogen, phosphorus, and cation content of the forest floor. Depending on the characteristics of the individual site, 1-3 forest floor horizons or strata may be sampled separately.

  • Nutrient content of mineral soil: three samples of the A horizon to a maximum depth of 7.5 cm at each of the litter measurement plots in the following manner: (1) one sample from the quadrant where the litter is removed and (2) two randomly selected from the other three quadrants. Samples will not be composited prior to chemical analysis for carbon, nitrogen, and macronutrient content.

Forest floor and mineral soil samples will be analyzed for total C and N content using a ThermoElectron Flash EA CNS Analyzer. Nitrate, ammonia, phosphate, and total phosphorus will be analyzed using a Lachat Quick Chem 8000 Flow Injector Analyzer, and cations (sodium, potassium, magnesium, and calcium) will be analyzed with a Dionex Liquid Chromatograph.



Literature Cited

Barger, R.L., and P.F. Ffolliott. 1969. Multiproduct timber inventory. Forest Products Journal 19:31-36.

Clary, W.P. 1964. A method for predicting potential herbage yield on the Beaver Creek pilot watershed. Pp 244-250 in Forage Plant Physiology and Soil-Range Relationships. ASA Special Publication 5. American Society of Agronomy, Madison, WI.

Ffolliott, P.F., W.P. Clary, and J.R. Davis. 1968. Some characteristics of the forest floor under ponderosa pine in Arizona. Research Note RM-RN-127. USDA Forest Service, Rocky Mountain Forest and Range Experiment Station, Fort Collins, CO. 4 pp.

Weatherspoon, P., and J. McIver. 2000. A national study of the consequences of fire and fire surrogate treatments: A proposal to the Joint Fire Science Program. Amended March 2000.


Go to the Southwest Watershed Science Team home page.

Project Contact: 

Principal Investigators:
Boris Poff - Rocky Mountain Research Station
Duncan Leao - Northern Arizona University